Nature Inspired Computation based Fractional Order I-TD and TID Controllers for Magnetic Levitation System

Abstract

In this paper, transient analysis of the Magnetic Levitation System (MLS) with fractional order feedback controllers has been investigated. MLS is an intricate open loop unstable system having various uncertainties. Fractional order controllers including Tilt Integral Derivative (TID) and its modified form Integral-Tilt Derivative (I-TD) have been optimized using two nature inspired computational techniques including Teaching Learning Based Optimization (TLBO) and Cuckoo Search Algorithm (CSA).  Further, four different performance indices such as ISE, ITSE, IAE and ITAE are utilized for the assessment of transient as well as steady state behaviour of the proposed controllers. MATLAB/Simulink has been used for the simulation purposes. The performance parameters of the system such as rising time (tr), overshoot (os), settling time (ts) and steady state error (ess) have been optimized with TID and I-TD controllers. Simulation results clearly reveal that proposed methodology efficiently control the open loop unstable Magnetic Levitation System (MLS)with optimum transient performance. Furthermore, it is evident from the results that TID controller gives superior performance in terms of settling and rising time. On the other end, I-TD controller gives less overshoot as compared to TID controller.
Keywords –Fractional Order controllers, Magnetic Levitation System (MLS), TID controller, I-TD controller, Heuristic or soft computation, Teaching Learning Based Optimization (TLBO), Open-loop unstable systems, Cuckoo Search Algorithm (CSA).